Advanced Microscopy Methods

• Novel interesting fluorophores and fluorescent

proteins

• Confocal Microscopy in general

• Spectral imaging (Emission Finger printing)

• Spinning disk confocal microscopy

• Multiphoton microscopy

• FRAP (Fluorescence Recovery After Photobleaching)

• FLIP (Fluorescence Loss in Photobleaching)

• FRET

• Superresolution Microscopy

Fluorophore development

• Traditional fluorophores:

fluorescein (FITC), rhodamine,

DAPI (4',6-diamidino-2-

phenylindole) for DNA

• newer dyes: Cy3, Cy5

• further developments: Alexa-

Dyes

• quantum dots: „fluorescent“

semiconductors; emission

wavelengths depends on the

size; one excitation wavelength

for several emission

wavelengths

bleaching

Interesting fluorescent proteins

fluor. protein Ex-Peak (nm) Em-Peak (nm) quantum yield comment

EBFP 380 440 0.18 Clontech

ECFP 433 (453) 475 (501) 0.4 Clontech

EGFP 488 507 0.6 Clontech

wildtype GFP 397 (475) 509 0.77 Aequorea victoria

EYFP 513 527 0.61 Clontech

Citrine 516 529 0.76 Griesbeck et al. 2001

DsRed 558 583 0.29 Clontech, tetramer

DsRed2 563 582 0.55 tetramer,

HcRed1 588 618 0.02 Clontech, dimer

PA-GFP

(Patterson

2002)

400 before act.

504 (397)

after

515 before act.

517 after act.

0.13

0.79

photoactivatable GFP, T203H mutant of

mammalian codon-optimized

wildtype GFP

PS-CFP 400 before act.

490 after act.

468 before act.

511 after act.

0.2

0.23

photoswitchable CFP, turns from cyan to

green after intense illum. at 405 nm

mOrange 548 562 0.69 Shaner et al., 2004

mStrawberry 574 596 0.29 Shaner et al., 2004

mCherry 587 610 0.22 Shaner et al., 2004

dTomato 554 581

0.69 Shaner et al., 2004, dimeric

New developments: photo-convertible

fluorescent proteins

pre-UV

exposure

after 30 sec UV

exposure

(350 nm, DAPI

excitation)

50% intensity

Dendra 2

0

20

40

60

80

100

120

-10 0 10 20 30 40

Dendra green

Dendra red

Photoconvertible fluorescent proteins

0

20

40

60

80

100

120

0 20 40 60 80 100 120

rel f

luo

r

sec

mOrange conversion to far-red (2x bleaching with 100% 488 nm in between)

mOrange

converted far red

control cell

mOrange

Far-Red

Photo-switchable fluor. protein Dronpa

0

20

40

60

80

100

120

-20 0 20 40 60 80 100 120 140 160 180 200

% o

f in

itia

l flu

or.

sec

start bleaching at 488 nm

start reactivation at 350 nm

Confocal Laser Scanning Microscopy

Photomultiplier

confocal pinhole

dichroic mirror

Objective

z-Motor

Scanner

Laser

Laser (Light fiber to the microscope

Detector

Modern Laser Microscopes

Photomultiplier

confocal pinhole

dichroic mirror

Objective

z-Motor

Scanner

Laser

Laser (Light fiber to the microscope

Detector

Confocal microscopy removes the blur

from thicker objects

http://zeiss-campus.magnet.fsu.edu/tutorials/opticalsectioning/confocalwidefield/index.html

Optical sectioning and 3D-

projections

10

Acquisition of a „z-stack“ (image slices along the z-axis)

allows reconstruction of a 3D-projection, which can be

shown as animation

z-stack

3D rendering projection

11

Spectral imaging

• „Emission finger printing“: emission scan of a microscopy sample („lambda

stack“ of images) at a given excitation wavelength (e.g. with Zeiss LSM

META systems or with Leica confocal microscopes…)

• Excitation scan (at a constant emission wavelength; e.g. using a

monochromator light source)

• Appropriate filter sets with different excitation and emission

(e.g using filter wheels)

• Increases the number of markers to be measured in parallel

• Can be used to discriminate fluorophores with overlapping spectra

• Can be used to discriminate specific fluorescence from autofluorescence

Resolving spectral information on a pixel-by-pixel basis

Zimmermann et al. (2003)

Sample with overlapping

fluorophores

Emission curves separated

into 8 channels (left) or

2 channels (right)

Equation matrix for the

channel signals based on

reference intensities in the

channels (GFPn and YFPn)

and the unknown

contributions of the

fluorophores

Unmixed fluorescence

(pseudo-coloured)

Example for Emission Fingerprinting

on a Zeiss LSM510 META: Separation

of GFP and YFP

Acquisition of a

reference lambda

stack for the first

fluorophore (GFP)

500 510 520 530 540 550 560Emission wavelength (nm)

0

50

100

150

200

250

Intensity

Obtain the spectral emission curve for the first fluorophore and

repeat the procedure for the second fluorophore

GFP

500 510 520 530 540 550 560Emission wavelength (nm)

0

50

100

150

200

250

Intensity

YFP

Unmixing of a mixed sample

(GFP-Actin and YFP-membranes)

Emission stack

Unmixed image

Example of

excitation scanning

spectral imaging

Excitation scan spectral imaging

0

0,2

0,4

0,6

0,8

1

1,2

450 460 470 480 490 500 510 520

nm

rel.

flu

or

YFP5

GFPc

> Analysis-Plugin available for ImageJ

(using a Polychrome IV monochromator

for excitation)

17

„Realtime“ confocal microscopy, Spinning disk

confocal microscopy (with Nipkow-disks)

gentle scanning (less bleaching

> good for sensitive life cells

Detection of the signal with a CCD-

camera

http://zeiss-campus.magnet.fsu.edu/tutorials/spinningdisk/yokogawa/index.html

Multiphoton Laser Scanning-Microscopy

A quantum physical phenomenon is used: at very high light

densities (using pulsed lasers, about 900 nm infrared light)

packages of 2 or more photons occur (just in the focal plane !).

These have the same energy as single photons of higher energy

(shoerter wavelength, e.g. 450 nm). Thus these photon packages

can excite a fluorophore, which emits then at for instance 520 nm

(mitted wavelength is horter than the excitation light wavelength !).

An important advantage is that the 900 nm light has a mucher

deeper penetration into tissue (approx. 1 mm), while conventional

excitation can image just down to 0.25 mm. Another advantage is a

reduced overall bleaching effect, as excitation photon packages

occur just in the focal plane.

19

Multiphoton Laser Scanning-Mikroskopie II

conventional excitation

(1-Photon > cone of

ecitation light)

2-Photon excitation:

only a spot of

excitation

20

FRAP: Fluorescence Recovery After

Photobleaching

to determine diffusion and mobility of molecules

y = span (1-e-kx) + bottom

inverse FRAP with novel fluorescent

proteins

22

Protocol: FRAP analysis on Zeiss LSM510

• Capture an image of the whole cell before bleaching

• Define a bleaching / scan region (and maybe in addition another scan region that is not bleached)

• Perform a time series with 1 scan prebleach, about 70 iterations of bleaching with 100% laser power and then 50-100 scans of the bleach region (and also the non-bleached control region if you specified one)- a good time resolution can only be obtained if just the small bleach region (and maybe the control region) is scanned - and not the whole cell; averaging of 2 or 4 scans reduces the electronic noise and leads to better quantifications.

• Capture an image of the whole cell after the FRAP time series (with the same conditions as the prebleach image – for calculating the total loss of fluorescence.

• If you want to save disk space: extract the FRAP region and save just this region instead of the whole image

• It is recommended to use the WCIF version of ImageJ for analysis: You can open the LSM-files with the built-in feature (which also allows opening the time values of the image series). Measure the mean fluorescence in a control region or for the whole cell for both the prebleach and the postbleach images and calculate the loss of overall fluorescence due to the bleaching in the region of interest (this is necessary for obtaining correct recovery values for the bleach region).

• Import the FRAP-image sequence, define a measurement region and apply the „intensity versus time plot“ plug-in – this will draw a graph of the FRAP curve; clicking the list button, shows a list of the numerical values (the first 4 parameters are dimension and position of the region, the rest are the fluorescence intensity values).

• Copy the fluorescence raw data from the list to the corresponding column of an Excel template

23

bottomespany kx )1(

• Calculate the difference of mean fluorescence from the background and

normalize the fluorescence values to 100% for the initial fluorescence.

• Divide the percent values by the correction factor calculated from the total

loss of fluorescence (e.g. if total fluorescence decreased from 1 to 0.9 then

divide the mean fluorescence of the FRAP regions for each time value by

0.9 to compensate for the loss in total fluorescence). A similar

compensation can be obtained by normalizing the FRAP fluorescence

values to the control scan region that was not bleached. This method also

compensates more exactly for the bleaching effect in the course of

scanning of the time series (this scanning-dependent bleaching effect is

opposed to the recovery of fluorescence in the bleach region due to

diffusion of non-bleached molecules in the bleach region). This “dynamic

correction” gives a somewhat better estimation of the curve (and the

kinetics of the recovery) – but leads in principle to results that are very

similar to the curve obtained with the “constant correction factor” (by

calculating the total loss in fluorescence based on the intensities of the

images that were captured before and after the FRAP-time series)

• For non-linear regression analysis (curve fit of the data to a single

exponential association algorithm): Copy the data to a fitting program

(such as GraphPad Prism) and perform the fitting with a “bottom to span”

algorithm:

FLIP: Fluorescence Loss in Photobleaching

… to determine the dynamic shuttling of molecules between different

compartments of the cell

A certain compartment A is repetitively bleached by the laser – and the fluorescence decrease in a different compartment B is monitored. Molecules that shuttle from B to A are bleached in A > thus the compartment B gets dimmer when there is a dynamic distribution of molecules between A and B.

nuclear FLIP(bleach in nucleus outside nucleoli)

0 100 200 300 400 500 600 7000

25

50

75

100

125nuclear

nucleolar

sec

FCS: Fluorescence Correlation Spectroscopy

… to determine diffusion coefficients and interactions between molecules.

The sample is illuminated by the laser at a very small spot, the movements of

molecules in this spot (in and out) cause fluorescence fluctuations, which are

analyzed by correlation functions

Ratio of donor emission and acceptor emission at the excitation

wavelength of the donor

ECFP and EYFP-Scans

0.0

0.2

0.4

0.6

0.8

1.0

1.2

350 400 450 500 550 600nm

rela

tiv

e f

luo

res

ce

nc

e

EYFP (Em)

EYFP (Ex)

ECFP (Em)

ECFP (Ex)

excitation

emission-1 emission-2

Limitations:

•concentration dependent

•controls are difficult

•donor and acceptor have to

colocalize completely (100%)

> just useful for FRET-biosensors

with covalent linkage between

donor and acceptor

image = Emission2 : Emission1

2 Filter-FRET Microscopy (Ratio Imaging)

27

Example for Ratio Imaging FRET-

Microscopy

3 Images are taken (under constant camera settings):

1. CFP (CFP-excitation and emission),

2. YFP (YFP-excitation and emission – this signal is not affected by FRET

3. FRET-Filter (CFP-excitation and YFP-emission).

A normalized FRET signal (image) can be calculated by using correction factors

obtained with single stained samples:

FRETN = IFRET - corrCFP x ICFP – corrYFP x IYFP

corrCFP : ca. 0.59 corrYFP : ca. 0.18

CFP / YFP neg. control CFP-YFP pos. control

3-Filter FRET Microscopy

Donor recovery after acceptor bleaching: An image of the donor in the presence of

the acceptor is taken, then the acceptor is bleached (partially), followed by

acquisition of a second donor image

ECFP and EYFP-Scans

0.0

0.2

0.4

0.6

0.8

1.0

1.2

350 400 450 500 550 600nm

rela

tiv

e f

luo

res

ce

nc

e

ECFP (Em)

ECFP (Ex)

excitation

Donor Image 1

bleaching of acceptor

Donor Image 2

EYFP ECFP

436 nm

476 nm

EYFP ECFP

436 nm

FRET

476 nm

FRET Microscopy by acceptor bleaching and

monitoring donor recovery (do not use for CFP / YFP !)

pos. control

(fusion protein)

neg. control

sample:

2 interacting

signaling

molecules

CFP Image before YFP-Bleaching after YFP-bleaching

Ratio Image

(pseudo colored)

Advantages:

concentration independent; donor and acceptor do not have to

colocalize; internal control; Ratio-image (after Acceptor-

bleaching/before Acceptor-Bleaching) = FRET-image

Example for a „Donor Recovery“ FRET-Microscopy

Comparison between the 3-filter method and

the donor recovery method

DRAP

(Donor Recovery After

Acceptor Photobleaching)

3-Filter corrected FRET image

(according to Youvan)

Normalization of FRET signals for concentrations

YJab1 HA-IKK2 FRET channel raw data

corrected FRET image

(not normalized for conc.)

FRET efficiency image

(normalized for conc.)

NFRET = (FF – d DF – a AF) / (DF AF)1/2 FRETc = (FF – d DF – a AF)

FRET analysis of protein interactions in patient

material

0

20

40

60

80

100

120

AR HDAC1 FRETc

flu

or.

normal mean

PIN mean

CA-Gl3 mean

> Protein interactions

of signaling

molecules can be

quantified and

visualized in tissue

sections

… this is slowed down in presence of a FRET acceptor

time series of images

EYFP ECFP

436 nm

FRET ECFP

436 nm

time series of images

476 nm 476 nm

CFP-Protein alone

CFP- and YFP-Protein

FRET Microscopy by analyzing the

kinetics of donor bleaching

offset.A0 ktey

single exponential decay

y... Fluor. Signal

A0... starting signal

k... decay constant

t... time

offset... final value

Fluorescence half time Tau: t = 0.69/k

FRETeff. E = 1 - (t without / t with Akzeptor.)

Probe mit FRET

Probe ohne FRET

advantages:

concentration independent

donor and acceptor don‘t have to

colocalize completely

Limitation: requires external control,

difficult to obtain a FRET-image

Donor-bleaching

kinetics

The Secret Nature of Some Fluorescent Proteins

Markedly Faster Bleaching in Commercial Mounting Fluids

and Photoactivation at YFP-Excitation

Attempts to perform acceptor-bleaching FRET microscopy fail in

fixed, mounted samples

Differences in bleaching kinetics in different

mounting fluids

Photoactivation of cyan fluor. Proteins upon

YFP-excitation (Ex500-514)

80

85

90

95

100

105

110

115

120

% o

f in

itia

l fl

uo

res

ce

nce

CFP

Cerulean

TFP

Not only CFP, but also Cerulean

shows this phenomenon

FRET microscopy by bleaching of acceptor and

monitoring the donor increase can lead to artefacts

Superresolution Microscopy I

STED: Stimulated Emission Depletion

A second laser (depletion laser) „trims“ the excitation spot (point-spread function, PSF) to a

smaller size. Resolution appr. 80 nm.

http://zeiss-campus.magnet.fsu.edu/articles/superresolution/introduction.html

Superresolution Microscopy II

Structured Illumination Microscopy (SIM)

A known pattern is projected into the

image plane at different angles and

interferes with sample structures,

creating Moiré pattern. Superresolution

information can now be captured by the

microscope from these structures by

mathematical algorithms.

(from www.zeiss.de)

normal image SIM image

Superresolution Microscopy

- by single molecule detection

STORM: Stochastic Optical Reconstruction Microscopy using single fluorescent molecules

PALM: Photoactivated Localization Microscopy

Resolution: appr. 30 nm, based on statistical calculation of the center of a Caussian Fit of a

single molecule. Requires a sensitive camera (e.g. EMCCD: Electron-multiplying charge-

coupled device cameras) – and some software, but no specific hardware

http://zeiss-campus.magnet.fsu.edu/articles/superresolution/introduction.html